INTRODUCTION At best, seismic exploration reveals only the variation of elastic velocity at differing depths in the earth. It remains to interpret this variation, usually in terms of the distribution of material with depth, occasionally in terms of the vertical variation of temperature, state, porosity, and other velocity-affecting parameters. Rarely can a unique interpretation of material distribution be made from seismic data alone. This is due to the large range of velocities even for well defined materials, and the common overlapping of velocity ranges of different materials. Velocity depends on a large number of factors. Among these are mineralogical composition, fluid content, temperature, pressure, grain size, cementation, direction with respect to bedding or foliation, and alteration. Lithologic classification is not at present fine enough to significantly reduce the velocity range for a given rock type. Sampling errors as much as 10 per cent for both laboratory and field determinations may occur. For these reasons a compilation of seismic velocities serves to provide known ranges (with no assurance that the possible extremes have been covered), and if possible, a statement of average values or typical ranges. Many field and laboratory studies have been devoted to the determination of seismic velocities. A complete listing is not attempted here, priority in selection being given to recent measurements on well described, common rocks. Field measurements are based almost entirely on explosion-generated waves in the frequency range of 10–100 cycles/sec. Laboratory determinations are typically made on small cylindrical samples. Resonance methods using audio frequencies,